1. Field of Invention
The invention relates to multiple phase electric induction machines, motors and generators. More particularly, the invention relates to an induction motor or generator that operates at a very high Power Factor over a broad range of loads without sacrificing important motor characteristics, such as efficiency and torque, while even improving some of these characteristics.
2. Description of Related Art
A three-phase AC Induction Motor is the primary motor type used in industrial applications. Induction motors consume 30% to 40% of the generated electrical energy worldwide, and up to 70% in certain developed nations. Because of their widespread use, the energy consumption parameters of induction motors are strongly regulated. Some of the parameters are controlled by particular standards. For example, motor efficiency is well appreciated, discussed, and controlled by the standards EPACT 92 and EISA 2007-MG1 Premium.
Another critically important energy consumption parameter is Power Factor (“PF”). PF is defined as the ratio of the amount of real power, in kW, consumed over the amount of apparent power, in KVA, consumed by a load. The combination of efficiency and PF defines the amount of current that a motor draws while it is running and producing a particular shaft power. The PF of induction motors has not yet been regulated by United States federal standards, even though induction motors are the primary creator of the reactive power demand, measured in units of Kilo-Volt Amperes Reactive (“kVAR”), in most electric utility systems. The most significant source for reactive power demand as kVAR for most, if not all electric utility systems, is created in large part by induction motors with poor PF. Many utilities charge a penalty to industrial consumers or similar consumers if the consumer's demand for reactive power exceeds certain thresholds. Many of these electric utilities penalize operators of motors with PF less than 0.95 to 0.85.
Another incentive to keep the PF of induction motors as high as possible is the effect of reactive power on transmission and distribution line losses. These losses are proportional to the square function of the total current which is in inverse proportion to the PF. The minimum threshold PF that a utility will tolerate from its customers is typically set by the utility. Minimum threshold PF values may vary from utility to utility, and by load and time of day.
TABLE 1Power FactorPower (hp)Speed (rpm)½ load¾ loadfull load0-518000.720.820.84 5-2018000.740.840.86 20-10018000.790.860.89100-30018000.810.880.91
A typical minimum threshold PF is 0.95. However, typical PFs of induction motors are significantly lower than 0.95 as shown in Table 1. Table 1 is taken from the website The Engineering Toolbox and is titled “PF for a Three Phase Electrical Motor,” which is available at http://www.engineeringtoolbox.com/power-factor-electrical-motor-d—654.html. As can be seen in Table 1, PF varies with the motor load, and in most cases with modern induction motors PF declines when a motor is used below its full load parameters. Further, most induction motors operate at less than full load.
Today, the most common method of PF correction is to utilize a capacitor bank installation at an entrance point to a large plant or at a utility company's desired location. The cables between a source transformer and induction motors carry reactive power. The reactive power of the cables will increase losses in a cable in reverse proportion to the PF of a given induction motor. For example, 80% PF means 44% higher cable losses than at Unity PF. These cable losses can be very significant in comparison with motor's own losses and therefore improving motor PF is very important for overall energy savings. Therefore, it is important to not just consider a motor's own efficiency, but overall system efficiency where cable losses are included. Therefore system efficiency is defined in this document as motor output power divided by sum of motor input power plus cable losses.
As can be seen by the shaded arrow in FIGS. 1a and 1b, active power, or real power, consumed is the same in both cases shown in FIGS. 1a and 1b. In order to correct the PF of the motor shown, a local capacitor may be added as compensation. Adding the capacitor allows for more real power to be provided by the system in FIG. 1b. 
There are several issues with the method of supplying reactive power to a motor through a local capacitor, including: 1) A transient current from the capacitor that is directly connected with the motor can trip the protective devices of the motor, especially since the full load settings of a compensated motor would be reduced versus when the motor is non-compensated. 2) Safety hazards may result from self-excitation of an induction motor, with medium to high inertia loads, immediately after the motor is de-energized (an additional relay switch is required to avoid these problems). 3) Capacitors directly connected to main power lines can multiply harmonic distortion issues because capacitor impedance for high harmonics is lower than for the fundamental frequency and harmonic issues are common with rectifiers or with variable speed drives. To solve this problem an expensive filter may be needed to correct these issues. 4) The variability of PF, especially with constantly changing motor loads, results in a need for consistent monitoring and adjustment of kVAR provided by the capacitors. The type of equipment to provide this monitoring and adjustment can be very expensive. 5) In the event that a local capacitor fails there may be a short circuit to ground. A short circuit of this type can result in a catastrophic failure for a capacitor bank installed in a utility application or in a locally installed application such as the one shown in FIG. 2.
Moreover, as the bulk of industrial motors are becoming more and more efficient the PF correction of motors becomes more critical. A newer, premium efficiency motor typically has a lower PF than the same size motor (in horsepower) of lower efficiency that is being replaced. If this trend is not addressed it may result in unintended consequences, including lower overall system efficiencies. Additionally, more capital improvements may be required due to necessary increases of cable sizes at plants because of the need to carry increased amounts of current required by motors with a low PF.
Some inventors have claimed to have solved the shortcomings of low PF and motor efficiency with a motor design having dual windings in a stator. One of the first examples of this technology is U.S. Pat. No. 4,446,416 to Wanlass, granted May 1, 1984, which is entitled “Polyphase Electric Machine Having Controlled Magnetic Flux Density.” Wanlass suggests that by maintaining the magnetic flux density in a stator the foregoing disadvantages regarding efficiency and PF will be overcome. Wanlass discloses a stator core having main windings and additional control windings. In a single phase application, the main windings are connected in series with capacitors and in parallel to the control windings. This means that the control windings have a direct connection to the power source and the main windings and capacitors are connected in parallel to the control windings.
More particularly, in a polyphase example, Wanlass teaches that a main stator winding that is wound on a magnetic core and includes a plurality of main and control windings. Each winding represents a single phase of a polyphase system. Capacitors are connected in series to each of the main windings and are used to reduce reactive power demand of the motor. Wanlass claims that the flux density is optimized in a polyphase machine by controlling the flux density in the stator core with the windings.
Embodiments are disclosed with regard to the mutual superposition of the control and main windings in Wanlass, along with the resulting PF and efficiency for these embodiments. In addition, the Wanlass '416 patent provides data on power of the windings in relation with the motor load. Power is defined as electric power of certain winding on its terminals without consideration of core losses and mechanical losses of the motor. In the '416 patent an example of a 1 hp, 230 volt 3 phase electric motor having a 10 μF capacitor installed in series with each main winding is provided. Table number 2 below, has been made based on the data in Wanlass' patent found in column 10, lines 5-32.
TABLE 2Power ofOutput ofControlPowerPowerMotor asWindingofOutputShaftPercentageasControlPower ofofPowerof FullPercentageWind-MainMotorofLoadMotorof Totaling (inWinding(inMotorRating ofEffi-Power ofWatts)(in Watts)Watts)(in hp)MotorciencyMotor 368 W479 W847 W1100.00%88.00%43%−174 W504 W326 W.34134.10%78.00%53%(negative)−390 W5010 W 114 W0575.70%37.30%325% (negative)For each of these outcomes the PF of the motor varied between 0.90 and 0.97.
Table 2 shows that with improper power-balance between the “main” and “control” windings, one of the windings acts as an internal brake to the motor. The power of the “control” winding, which is the winding that is directly connected to the power source, is negative in some instances. The '416 patent describes this situation as the “control” winding generating power, however, if a winding in an electric machine while its motoring operation is “generating,” this actually means the winding is acting as internal brake. If a winding is acting as an internal brake, as the control winding does in the results shown in Table 2, this means that the other windings are working harder than necessary and are producing additional losses while delivering excessive power. When one winding is delivering too much power and the other is acting as a break, this is definitely not an optimum power-balance and this explains why the Wanlass design motor efficiency suffers.
Prior publications and actual physical tests point out a 2-4% efficiency reduction in the Wanlass design vs. the standard machine under light loads, and either no or slight efficiency gains under full load. (page 65, Baghzous and Cox 1992, and my FEA analysis). Furthermore, Wanlass' commentary in his patent displays his design having suboptimal power balance and significant breaking effect (column 10, line 5-32). Similar conclusions have being reached by other experts in the discussion section of the paper by Umans and Hess (1983), for example the FIG. F3 on page 2923.
In addition, another design for a motor with dual windings that was said to improve upon the winding disclosed by Wanlass is the “Unity Plus” winding method. The “Unity Plus” winding method is described as one of the embodiments of U.S. Pat. No. 4,808,868 ('868) to Gary D. Roberts described in the patent as a floating quasi-resonant concepts, however in publications it is almost always referred to as “Unity Plus” method. The other embodiment in the '868 patent is a “quasi-double resonant circuit” and we refer to this as Roberts throughout the document. The Unity Plus winding method includes two windings both of which are placed in the same stator slots. A first winding is connected to an electrical source and is located above a second winding, that is not connected to the electrical source, in the slot of the stator. Capacitors are mounted to the motor externally and are connected in parallel to the second winding.
Despite the claims made by those who advocate for the Unity Plus winding method as a way to increase motor efficiency, more than one author has noted that those claims seem dubious. In his paper titled “Unity Plus Motor Winding Method Advantages and Disadvantages,” Donald Zipse states that “Some of the data offered by Unity Plus has indicated greater than 100 percent efficiency, which at this point in time, physical science has not achieved the perpetual motion machine. From this, one can assume the data is incorrect.” (page 118, Zipse 1990). Zipse also describes how the Wanlass winding design failed to meet its claims of improved efficiency and the Unity Plus winding method came along in the 1980's to fill this void. Zipse states that the Unity Plus winding method was claimed to provide higher efficiency than the Wanlass winding, have close to a unity PF, and have higher starting and breakdown torque than the Wanlass winding, among other things.
Zipse found that the Unity Plus winding method did improve PF mostly due to the addition of capacitors with the motor windings. Despite the PF improvement, Zipse was clear to point out that a higher PF does not equal higher efficiency. Further, Zipse provides an example based on manufacturer reports that a motor that was rewound using the Unity Plus winding method actually had a lower efficiency than the original motor. Moreover, as cited in the Zipse paper, the amount of additional copper wire that was used by Unity Plus and Wanlass in some machines was as high as 59%.
Additionally, in the paper “Efficiency of Dual Winding Induction Motors With Integral Capacitors” by Y. Baghzouz and M. D. Cox, a standard electric motor was compared to a motor that had been rewound according to the Unity Plus winding method (in this article Unity Plus and Roberts are used interchangeably) and the Wanlass designs. Baghzouz and Cox ultimately concluded that the Unity Plus design did not improve efficiency over the standard motor. In fact, the efficiency of the motor proved to be worse over its entire operating range. They found this to be true even when the capacitor size was optimized for a given load. In addition, Baghzouz and Cox found that while both the Unity Plus method design and the Wanlass design achieved near unity PF. They said this could have been accomplished merely by adding shunt capacitors at the stator terminals without affecting the performance of the motor. Further, prior publications and the tests carried out in these publications point out a 2-4% efficiency reduction in the Wanlass design compared to a standard machine under light loads, and either no or slight efficiency gains under full load.
Further, another dual winding motor design is taught in U.S. Pat. No. 7,034,426 ('426 patent) and U.S. Pat. No. 7,227,288 ('288 patent) issued to Gerald Goche. The '426 and '288 patents describe similar motors. The '426 patent discloses an electric motor that is single or multiphase, with at least three phases, and has at least two poles. The electric motor also includes main windings and additional or de-saturation windings that are fed by at least one capacitor. Each additional winding is fed through a capacitor at a different phase angle and at a different field direction than each respective main winding. In addition the total cross section of wire used on each main and additional winding is a predetermined ratio. That ratio is approximately two-thirds for a main winding and one-third for an additional winding.
The '288 patent teaches that an electric motor has a main winding and an additional winding that are wound from different size conductors. The first conductor used to form the main winding has a greater size than the conductor used to form the additional winding. The number of turns of the additional winding is at least equal to half of the number of turns in the main winding. The number of turns in the additional winding may be equal to but not more than the main winding.
These patents claim to improve both the PF of the motor over a full range of loads and the efficiency of the motor. The '288 patent states that it teaches a motor that will cause a substantial drop in the starting and operating currents of the motor across all loads. Despite these claims, test results indicate that a motor configured in the fashion described in the '288 patent does not meet these claimed goals. As discussed below, a Goche design motor exhibits reduced motor efficiency at 100% of its rated load and this may lead to significant overheating and premature failure.
Furthermore, in the '288 patent, Goche describes the interaction between additional windings and main windings as being that which produces the undesired internal braking, similar to what we described above in reference to the Wanlass' design, “The first and second potential lines are electrically connected in parallel relation to the first and second main windings. The first and second additional windings generate a field in opposite direction relative to the field of the first and second main windings, respectively.” According to this statement the power of an additional winding would be in an opposite direction of a main winding, and hence the power of one of the windings would be negative. In fact, whereas the power of the windings of Wanlass were only braking at light loads, as described above, the Goche auxiliary windings are generating, or the power is negative, at all loads.
As can be seen above, both Goche and Roberts fail to recognize the relationship between the power of a set the windings compared to the overall power of both sets of windings in a dual winding motor. Wanlass did provide an example of generation of one winding in a dual winding machine, however he failed to recognize the negative effect of this generation on motor performance. Furthermore, Wanlass' commentary in the '416 patent displays the significant braking effect that generation has on the efficiency of the motor. Similar conclusions have been reached as discussed in the article by Umans and Hess referenced above. Accordingly, there is a need for an induction motor design that would reduce motor cable losses by minimizing kVAR requirements, which is achieved with a high PF, without sacrificing, and even improving upon, efficiency and torque characteristics in both light and heavy loads as compared with standard motor design.